Vegalite MCP Server
Vega-Lite data visualization and table storage for AI-driven analytics workflows.
Data last scanned today · Reviewed today
Overview
The Vegalite MCP Server provides LLMs with a direct interface for data visualization and state management. It allows users to store data tables through a save_data tool and subsequently generate visualizations using the Vega-Lite specification language. Users can configure the output to return either raw JSON specifications as text or base64 encoded PNG images. Integration involves configuring the server command with specific output types, such as png or text, to suit the requirements of the host environment. This tool is designed to bridge the gap between abstract model data and rendered visual representations.
Our verdict
This server carries a high-risk score of 20/100, primarily due to the lack of a declared license and inactivity, as indicated by the fact that the last commit was 480 days ago. Because the repository is community-maintained and has not undergone editorial review, it is best suited for experimental environments where dependency supply-chain security is not a primary concern.
- Zero direct dependencies reduce potential supply-chain vulnerabilities
- Supports both text-based JSON artifacts and PNG image rendering
- No license detected creates legal ambiguity for adoption
- Stagnant development with no updates for over 480 days
- Security configuration and authentication methods are unreviewed
Tools
| Tool | Description | Risk |
|---|---|---|
| save_data | Save a table of data agregations to the server for later visualization | medium |
| visualize_data | Visualize a table of data using Vega-Lite syntax | low |
Compatibility
| Client | Local | Docker | Remote | Read-only |
|---|---|---|---|---|
| Claude Desktop | ||||
| Cursor | ||||
| VS Code | ||||
| Windsurf | ||||
| ChatGPT |
Frequently asked questions
›What output formats does the visualize_data tool support?
The server supports either text or png output types, which can be configured using the --output_type flag.
›How do I make data available for visualization?
You must first use the save_data tool to store your table of data with a unique name before it can be referenced in a visualization.
›What information is required to visualize a saved dataset?
You need to provide the name of the saved data table and a valid JSON string containing the Vega-Lite specification.
›How is the server typically configured for Claude Desktop?
The server is configured by adding it to the claude_desktop_config.json file using the uv command with the absolute path to the server directory.
Alternatives
Badge
Maintain this server? Add the live badge to your README.